AI RESEARCH
Strategic Heterogeneous Multi-Agent Architecture for Cost-Effective Code Vulnerability Detection
arXiv CS.LG
•
ArXi:2604.21282v1 Announce Type: cross Automated code vulnerability detection is critical for software security, yet existing approaches face a fundamental trade-off between detection accuracy and computational cost. We propose a heterogeneous multi-agent architecture inspired by game-theoretic principles, combining cloud-based LLM experts with a local lightweight verifier.